Modding Azur 640A

Member
Joined 2010
Paid Member
.... what in this amp consumes 10w at standby?.....
There is no separate transformer to manage standby mode in this model or the similar 540A. The standby mode simply cuts AC from the main power transformer secondary windings to the power amp rectifier, leaving the low voltage supplies for preamps, probably the headphone amps, (12V trigger?) 24V Abus control and 5V microprocessor still operational, so they can respond to wake up commands and switch the power amplifiers and protection circuits back on.

No doubt, newer models will conform to lower standby current regulations and use a different design approach. I don't think converting that to a separate standby transformer for the 5V supply would be easy or work correctly without a design re-think.
 
Standby transformer of dubious quality?

Do they keep the amp's front end powered up to keep it 'warm'?

Switch it off with a proper inline switch or switched power board. 10w in standby is either a fault or very poor design.
looked at this and this is definitely poor design. there are two stages, pre and power amp, the pre got +-15ac power from main transformer low voltage secondaries and change it to +-15dc and +5dc, and after that the latter powers standby to provide signal to high voltage secondaries relay switch which keeps them off until ir or manually switched on. there is no relay switch for keeping pre amp off as voltage change is done on it (and perhaps protection stuff and standby). there is missing small separate transformer to deliver power for standby only and ir and to keep main transformer full off.

there is service manual available (540a v2) which covers also higher rated 640a v2.

any hints how good design solves it appreciated.

tapnięte talkiem
 
thanks Ian for reply. seems we conclude to similar results.

to fix it one need to add relay switch at supply side to cut off low secondaries in similar way add it is done for high line and then use separate transformer to supply +5vdc line in pre amp. if the controller chip is 5vdc then it might succeed. the +5vdc converter from pre amp needs to be perhaps cut off to not cause problems when new relay restores supply for pre amp. based on SM do you think it is doable?



tapnięte talkiem
 
ok, I've cut if 5v rectifier and connected instead over supply from mobile charger. it seems to work in standby with disconnected 15vdc until power switch it's turn on, then protection led is blinking, perhaps due to no power in amp section after while. the power consumption at standby is now 3w - 10w lower than base. now I have to rethink what to do with existing high voltage relays and new one for low power. the existing ones are supplied by unregulated 17 vdc just after gretz bridge. there is standby signal from pic and this unreg to drive relay coil. i thinking about changing existing relays to 5vdc type so i could replace this unreg branch with 5.6vdc from new supply. or alternatively add one 5vdc relay and let it turn on low power only, then ??? high power relays will triggers or not. the question is is it enough time to have second relays bump until pic secure something is not normal... have to experiment. awaiting relays.

tapnięte talkiem
 
Well capacitors and opamps done on the preamp PCB. Noticed there are two 2W carbon resistors which are part of the speaker protection circuity which are looking nice and dark brown. Obviously see a bit of heat. I have some 5W wirewound ones here which I am looking at replacing them with. I can't see any issues in doing this but anyone no of any reasons why I shouldn't?
 
Well capacitors and opamps done on the preamp PCB. Noticed there are two 2W carbon resistors which are part of the speaker protection circuity which are looking nice and dark brown. Obviously see a bit of heat. I have some 5W wirewound ones here which I am looking at replacing them with. I can't see any issues in doing this but anyone no of any reasons why I shouldn't?
because it's just circuit to drive relays and have nothing common with audio signal path?

tapnięte talkiem
 
eco mod finished

finally i'd time to finish my mod. herein some pictures and explanation what have been done.

03412ac647046dca6fa101442fcedffb.jpg


wires from power amp to additional capacitors mounted side of board.

02d89d3e9aa8ee418b50b22333f2d36b.jpg


first sound check with additional capacitance and U7 op-amp exchanged to AD8599.

2250205ca7dc22453fcf176773c8fdcd.jpg


93c43ec2471fcf0d24fb9c36def80331.jpg


with mounted dac module TDA1543 on the left. the 2 rectifiers from dac are screawed to top of pre-amp heatsink.
voltage supply is taken from +15vdc link on pre-amp. output signal from dac is connected to tape monitor input
of pre-amp.

8b88eabb756f63365e1ac7dc5f68c553.jpg


back side with optical input added.

2a800925f6478bd8f0a62e577119e851.jpg


727207d088db221f52ca05af2ee24e6f.jpg


all poor xunda electrolytic capacitors has been exchange with some good quality ones

ecb5a061382728e766655393df257db2.jpg


ff2cd5b5c89dd186707beafd7071d49d.jpg


2*4*3300uf 63v mounted on side below trafo

d4ded9ae17f85a0669895db1e5f9e039.jpg


final look with additional relay on 17vac line to preamp. relay is driven by standby signal from lk105 and +5vdc taken from links on board.
additionaly I've disconnected D1 from +15vdc line and connected to +5vdc. the +5vdc power supply from mobile charger is in left bottom corner. it is conected in place of removed U3 rectifier on pre-amp board.

the only problem is with abus module standby signal. when connected the amp goes to protection.
when whole connector CN1 from abus is disconnected it starts ok but the remote control is not working.
so i've removed standby signal from CN1 connector (pin 2) and plug it in. now all works. i've checked module but have
not found any errors. maybe it is caused by too much current load on +5vdc line. the +5vdc power supply i've mounted is only 800mA capable.

the power consumption on stand-by after modifications is 3w compared to 13w before (11w + 2w additional from dac module).


tapnięte talkiem
 
the best from this mod is that I've no longer to plug in and out power cord every time I want to use the amp and can start it using irda on mobile phone then open spotify and play music to Amazon fire tv which is connected to new optical input in amp. the sound quality from dac is superb and very feasible to use.

tapnięte talkiem
 
Hi all,

I modded my 640A (v1) and 640C (v2). Below are the components I changed in the 640A:
Nichicon UKW 3300uf / 50V
Elna Silmic II 10 uF 25V
Elna Silmic II 47uF 50V
Elna Silmic II 47uF 16V
Elna Silmic II 100 uF 25V
Elna Silmic II 100uF 16V

Elna Silmic II 100 uF 35V
Elna Silmic II 220uF / 50V
Elna Silmic II 470uf / 35V
WIMA MKP10 100nF (used for OP-amps decoupling, as close as possible from pins)

WIMA MKP4 47nF
10uF NP Nichicon muse UES (bipolar electrolytic for DC decoupling)
Relay G2RL-24-DC12 (Had to change the output relay as there were some false contacts)
OP Amp LME49720NA

If anyone is still interested, I could provide 640A v1 service manual and the ref of components changed.
 
Update: I did step by step modding and listened between each steps.
I love the improvement from the new capacitors (gain in dynamic, bass seems to go deeper); But I am not really sure I like the OpAmp update: They sound too laid back, and kill the involvment with a too decontracted presentation. it sounds like bass and treble is well contoured, but mids somehow lack texture. I think that I might have been used to the NE grainy mids too much, but still, I did not expect such a big difference.
 
Hi All,

I though I will drop my experiences with this amp in case someone finds it useful.
Similar to other people in this thread I have done a quick swap of op amps to opa2134 or similar. And so i discovered that sound changed much more that I expected. Shortly after this I decided to connect my source directly to power stage and use DAC for volume control. That was the point of no return - no more op amps in my signal path! I know the dac will still have VI converter or at least a buffer:)
Change of sound reminded me a review of one of Pass amps that i read ages ago - "orchestra took a deep breath, sat back, and played again". it was a small step in that direction, while there was still plenty of things wrong with the sound.


Next big win was to push the amp i to class A as much as possible. Turning ports almost all the way up, resulted in drop of about 100mV on emiter resistors giving around 450mA of current. Good enough for my conditions with 8R nominal speakers to make sure it will sit in class A most of the time.
Radiator got a bit too hot, so I squeezed in a couple of 44mm (or 40) low profile fans on top of the radiator, wired them to 5V and still managed to close the top cover with 2-3mm clearance. From my sofa I could not hear the fans.

sound was kinda ok, less fatigue.

Another thing I have done at some point was to remove a couple of caps near "direct" switch and wire the switch contacts associated with "direct" position directly to a pair of RCA terminals at the back where my Dac was connected. Kinda mid way compromise between functionalty and sound. This kept one opamp in volume control, but still better than 3 or 4 of them.

So there you go, few things to consider if you are lazy like me and listen to mainly one descent quality source.
 
remove the connector(which connects to power amp) from preamp. directly connect it to the source and the amp sounds magical. the pre section in this amp is what makes it sound weaker at Lowe volumes. they have put in so many nee5532 in the signal path as buffers.
note:- while connecting the source signal make sure the ground is linked between both the channels as there is no ground present in one of the channels(they have given the ground from pre for one channel) or else it will hit the protection.

the amp sounds totally different. powerful and open.